Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431644
János Tapolcai, P. Fodor, G. Rétvári, M. Maliosz, T. Cinkler
In this paper an advanced framework for minimum interference routing is proposed, which, despite its universal nature, is best suited for traffic engineering in optical networks. Minimum interference routing is a method for engineering network resources to minimize the interference among high-volume traffic sources, in an attempt to maximize the useful throughput of the network. The main motivation behind the proposed framework Is a more advanced way of defining the "interference" between routes. This promotes the allocation of light-paths in an optical network without presuming any knowledge of future requests. The basic idea is that network links are classified into distinct criticality classes and additional capacity is defined for each criticality class. This additional capacity represents the amount of additional traffic that can be routed on those links, without increasing the network-wide interference. We provide a concise mathematical formula for class-based interference and propose a polynomial time heuristic to derive criticality classes. Based on our findings a flexible path selection scheme called MIRO is introduced. Finally, through extensive simulation studies, we show that the class-based interference framework more thoroughly describes link criticality than traditional methods, which transforms into more efficient routing in optical transport networks.
{"title":"Class-based minimum interference routing for traffic engineering in optical networks","authors":"János Tapolcai, P. Fodor, G. Rétvári, M. Maliosz, T. Cinkler","doi":"10.1109/NGI.2005.1431644","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431644","url":null,"abstract":"In this paper an advanced framework for minimum interference routing is proposed, which, despite its universal nature, is best suited for traffic engineering in optical networks. Minimum interference routing is a method for engineering network resources to minimize the interference among high-volume traffic sources, in an attempt to maximize the useful throughput of the network. The main motivation behind the proposed framework Is a more advanced way of defining the \"interference\" between routes. This promotes the allocation of light-paths in an optical network without presuming any knowledge of future requests. The basic idea is that network links are classified into distinct criticality classes and additional capacity is defined for each criticality class. This additional capacity represents the amount of additional traffic that can be routed on those links, without increasing the network-wide interference. We provide a concise mathematical formula for class-based interference and propose a polynomial time heuristic to derive criticality classes. Based on our findings a flexible path selection scheme called MIRO is introduced. Finally, through extensive simulation studies, we show that the class-based interference framework more thoroughly describes link criticality than traditional methods, which transforms into more efficient routing in optical transport networks.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121183541","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431652
J. Palmer, I. Mitrani
We consider the problem of how best to structure and control a distributed computer system containing many processors. The performance trade-offs associated with different tree structures are evaluated approximately by applying appropriate queueing models. It is shown that, for a given set of parameters and job distribution policy, there is an optimal tree structure that minimizes the overall average response time. This is obtained numerically through comparison of average response times. A simple heuristic policy is shown to perform well under certain conditions.
{"title":"Optimal tree structures for large service networks","authors":"J. Palmer, I. Mitrani","doi":"10.1109/NGI.2005.1431652","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431652","url":null,"abstract":"We consider the problem of how best to structure and control a distributed computer system containing many processors. The performance trade-offs associated with different tree structures are evaluated approximately by applying appropriate queueing models. It is shown that, for a given set of parameters and job distribution policy, there is an optimal tree structure that minimizes the overall average response time. This is obtained numerically through comparison of average response times. A simple heuristic policy is shown to perform well under certain conditions.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114506877","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431701
G. Albertengo, C. Pastrone, G. Tolu
The continuously increasing diffusion of mobile devices such as laptops, PDAs and smartphones, all equipped with enhanced functionalities, has led to numerous studies about mobility and to the definition of new network architectures capable to support it. Problems related to mobility have been addressed mostly operating on the network or transport layers of the Internet protocol stack. As a result, most of these solutions generally require modifying the TCP and/or the IP protocol. Although this approach is well suited to handle mobility, it lacks in compatibility with the Internet Protocol Suite. This consideration led us to study a fully TCP compatible and flexible approach we dubbed MOON, for MObile Overlay Network. This network architecture is currently under design at LIPAR, the Internet, protocols and Network Architecture Lab of Politecnico di Torino.
{"title":"MOON: a new overlay network architecture for mobility and QoS support","authors":"G. Albertengo, C. Pastrone, G. Tolu","doi":"10.1109/NGI.2005.1431701","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431701","url":null,"abstract":"The continuously increasing diffusion of mobile devices such as laptops, PDAs and smartphones, all equipped with enhanced functionalities, has led to numerous studies about mobility and to the definition of new network architectures capable to support it. Problems related to mobility have been addressed mostly operating on the network or transport layers of the Internet protocol stack. As a result, most of these solutions generally require modifying the TCP and/or the IP protocol. Although this approach is well suited to handle mobility, it lacks in compatibility with the Internet Protocol Suite. This consideration led us to study a fully TCP compatible and flexible approach we dubbed MOON, for MObile Overlay Network. This network architecture is currently under design at LIPAR, the Internet, protocols and Network Architecture Lab of Politecnico di Torino.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130568143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431685
Javier Munoz Kirschberg, M. Delgado, S. S. Ribes
This paper introduces a new protocol, called RCCMP (reliable congestion controlled multicast protocol). This protocol has been designed to be simple, scalable (NAK suppression), reliable and TCP-friendly. The congestion control is a central part of the protocol, where the feedback of the worse receiver is used to control a transmission window in a TCP-like fashion. The scalability issue is addressed with an exponential tuners scheme, that is also used to estimate the number of receivers involved in the communication. The protocol neither needs support from network elements nor maintains state information dependent of the number of receivers. The paper also presents a number of simulation results on its performance.
{"title":"RCCMP: reliable congestion controlled multicast protocol","authors":"Javier Munoz Kirschberg, M. Delgado, S. S. Ribes","doi":"10.1109/NGI.2005.1431685","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431685","url":null,"abstract":"This paper introduces a new protocol, called RCCMP (reliable congestion controlled multicast protocol). This protocol has been designed to be simple, scalable (NAK suppression), reliable and TCP-friendly. The congestion control is a central part of the protocol, where the feedback of the worse receiver is used to control a transmission window in a TCP-like fashion. The scalability issue is addressed with an exponential tuners scheme, that is also used to estimate the number of receivers involved in the communication. The protocol neither needs support from network elements nor maintains state information dependent of the number of receivers. The paper also presents a number of simulation results on its performance.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123981288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431686
A. Nogueira, M. R. D. Oliveira, P. Salvador, R. Valadas, António Pacheco
The (reliable) classification of Internet users, based on their hourly traffic profile, can be advantageous in several traffic engineering tasks and in the selection of suitable tariffing plans. For example, it can be used to optimize the routing by mixing users with contrasting hourly traffic profiles in the same network resources or to advise users on the tariffing plan that best suits their needs. In this paper we compare the use of discriminant analysis and artificial neural networks for the classification of Internet users. The classification is based on a partition obtained by cluster analysis. We classify the Internet users based on a data set measured at the access network of a Portuguese ISP. Using cluster analysis performed over half of the users (randomly chosen) we have identified three groups of users with similar behavior. The classification methods were applied to the second half of users and the obtained classification results compared with those of cluster analysis performed over the complete set of users. Our findings indicate both discriminant analysis and neural networks are valuable classification procedures, with the former slightly outperforming the latter, for the specific scenario under analysis.
{"title":"Classification of Internet users using discriminant analysis and neural networks","authors":"A. Nogueira, M. R. D. Oliveira, P. Salvador, R. Valadas, António Pacheco","doi":"10.1109/NGI.2005.1431686","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431686","url":null,"abstract":"The (reliable) classification of Internet users, based on their hourly traffic profile, can be advantageous in several traffic engineering tasks and in the selection of suitable tariffing plans. For example, it can be used to optimize the routing by mixing users with contrasting hourly traffic profiles in the same network resources or to advise users on the tariffing plan that best suits their needs. In this paper we compare the use of discriminant analysis and artificial neural networks for the classification of Internet users. The classification is based on a partition obtained by cluster analysis. We classify the Internet users based on a data set measured at the access network of a Portuguese ISP. Using cluster analysis performed over half of the users (randomly chosen) we have identified three groups of users with similar behavior. The classification methods were applied to the second half of users and the obtained classification results compared with those of cluster analysis performed over the complete set of users. Our findings indicate both discriminant analysis and neural networks are valuable classification procedures, with the former slightly outperforming the latter, for the specific scenario under analysis.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114403179","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431649
A. Gulyás, J. Bíró
The workload loss ratio (WLR) is a key parameter from the point of quality of service (QoS) provisioning in packet switched networks and it's estimation is an important issue. The existing results in the area of WLR approximation usually interpret the workload loss as a product of some well assessable quantities. We call this approach as the indirect approximation of the WLR. The drawback of this approach is, that each estimation has an error and the product of these errors could result in highly inaccurate bound. This work deals with the upper approximation of the workload loss ratio based on it's original definition and proposes direct bounds on the WLR. Besides the presentation of the new bounds an extensive systematic performance analysis of the results is also given, in which we show that the direct approach gives closer bound in most cases.
{"title":"Direct and indirect methods for packet loss approximation in buffered systems","authors":"A. Gulyás, J. Bíró","doi":"10.1109/NGI.2005.1431649","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431649","url":null,"abstract":"The workload loss ratio (WLR) is a key parameter from the point of quality of service (QoS) provisioning in packet switched networks and it's estimation is an important issue. The existing results in the area of WLR approximation usually interpret the workload loss as a product of some well assessable quantities. We call this approach as the indirect approximation of the WLR. The drawback of this approach is, that each estimation has an error and the product of these errors could result in highly inaccurate bound. This work deals with the upper approximation of the workload loss ratio based on it's original definition and proposes direct bounds on the WLR. Besides the presentation of the new bounds an extensive systematic performance analysis of the results is also given, in which we show that the direct approach gives closer bound in most cases.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125053982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431683
Yuming Jiang, P. Emstad, Anne Nevin, Victor Nicola, Markus Fidler
To provide statistical service guarantee and achieve high network utilization, measurement-based admission control (MBAC) has been studied for over one decade. Many MBAC algorithms have been proposed in the literature. However, most of them belong to aggregate MBAC algorithms which assume or require that (1) first-in-first-out (FIFO) is used for aggregating flows; (2) statistical service guarantees are provided to the aggregate of admitted flows; (3) each flow requires and experiences the same statistical service guarantees as the aggregate. In this paper, we focus on per-flow MBAC that aims to provide possibly different statistical service guarantees to individual flows in an aggregate. Particularly, we propose a simple per-flow MBAC algorithm in which dynamic priority scheduling (DPS) is adopted to aggregate flows. With this DPS-based per-flow MBAC algorithm, a newly admitted flow is always given a lower priority level than all existing flows, and its priority level is improved if an existing flow leaves the system. Consequently, once a flow is admitted, its received service will not be adversely affected by other flows admitted after it. Because of this, there is no need to re-check or adjust network resources allocated to existing flows due to the admission of a new flow.
{"title":"Measurement-based admission control for a flow-aware network","authors":"Yuming Jiang, P. Emstad, Anne Nevin, Victor Nicola, Markus Fidler","doi":"10.1109/NGI.2005.1431683","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431683","url":null,"abstract":"To provide statistical service guarantee and achieve high network utilization, measurement-based admission control (MBAC) has been studied for over one decade. Many MBAC algorithms have been proposed in the literature. However, most of them belong to aggregate MBAC algorithms which assume or require that (1) first-in-first-out (FIFO) is used for aggregating flows; (2) statistical service guarantees are provided to the aggregate of admitted flows; (3) each flow requires and experiences the same statistical service guarantees as the aggregate. In this paper, we focus on per-flow MBAC that aims to provide possibly different statistical service guarantees to individual flows in an aggregate. Particularly, we propose a simple per-flow MBAC algorithm in which dynamic priority scheduling (DPS) is adopted to aggregate flows. With this DPS-based per-flow MBAC algorithm, a newly admitted flow is always given a lower priority level than all existing flows, and its priority level is improved if an existing flow leaves the system. Consequently, once a flow is admitted, its received service will not be adversely affected by other flows admitted after it. Because of this, there is no need to re-check or adjust network resources allocated to existing flows due to the admission of a new flow.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116768192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431658
Kin-Hon Ho, M. Howarth, N. Wang, G. Pavlou, S. Georgoulas
This paper considers traffic engineering (TE) for the provision of end-to-end quality of service (QoS) guarantees across multiple autonomous systems (ASes) in the Internet. We review an inter-AS QoS management model, and consider algorithms for offline inter-AS and intra-AS TE. We consider two approaches, decoupled and integrated, to allow these systems to work together to achieve optimal TE performance. We quantify through simulation the lower TE costs incurred by the integrated approach.
{"title":"Two approaches to Internet traffic engineering for end-to-end quality of service provisioning","authors":"Kin-Hon Ho, M. Howarth, N. Wang, G. Pavlou, S. Georgoulas","doi":"10.1109/NGI.2005.1431658","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431658","url":null,"abstract":"This paper considers traffic engineering (TE) for the provision of end-to-end quality of service (QoS) guarantees across multiple autonomous systems (ASes) in the Internet. We review an inter-AS QoS management model, and consider algorithms for offline inter-AS and intra-AS TE. We consider two approaches, decoupled and integrated, to allow these systems to work together to achieve optimal TE performance. We quantify through simulation the lower TE costs incurred by the integrated approach.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115132290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431656
F. Ferreira, A. Pacheco
In this paper we address the simulation of stochastic processes ordered in level crossing, where, as proposed by Irle and Gani, X is slower in level crossing than Y if it takes X stochastically longer than Y to exceed any given level. This ordering has immediate applications to the analysis of first-passage times, which are important for detailed performance analysis of distributed computer and communication systems, like the response time of an Internet search-engine query. We provide algorithms for the simulation of level crossing ordered discrete time Markov chains, semi-Markov processes, and continuous time Markov chains under conditions derived recently.
{"title":"Simulation of semi-Markov processes and Markov chains ordered in level crossing","authors":"F. Ferreira, A. Pacheco","doi":"10.1109/NGI.2005.1431656","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431656","url":null,"abstract":"In this paper we address the simulation of stochastic processes ordered in level crossing, where, as proposed by Irle and Gani, X is slower in level crossing than Y if it takes X stochastically longer than Y to exceed any given level. This ordering has immediate applications to the analysis of first-passage times, which are important for detailed performance analysis of distributed computer and communication systems, like the response time of an Internet search-engine query. We provide algorithms for the simulation of level crossing ordered discrete time Markov chains, semi-Markov processes, and continuous time Markov chains under conditions derived recently.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127241915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431651
D. Barth, P. Berthomé, J. Fourneau, C. Laforest, S. Vial
We analyze the performance of a new routing strategy for all-optical packet networks. This strategy improves the Eulerian routing technique, a convergence routing based on a an Eulerian directed cycle. This new technique allows to use shortcuts along the Eulerian cycle. Usual Eulerian routing provides deterministic transport delays but has a very low network utilization. With this new strategy the average transport time is much smaller and we can prove a deterministic upper bound of the transport delay, unlike deflection routing which suffers from livelocks. We study the performance guarantees provided by this new algorithm using graph arguments for the ending property and simulations to give some insights for the performance of the algorithm.
{"title":"Performance evaluation of short-cut Eulerian routing","authors":"D. Barth, P. Berthomé, J. Fourneau, C. Laforest, S. Vial","doi":"10.1109/NGI.2005.1431651","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431651","url":null,"abstract":"We analyze the performance of a new routing strategy for all-optical packet networks. This strategy improves the Eulerian routing technique, a convergence routing based on a an Eulerian directed cycle. This new technique allows to use shortcuts along the Eulerian cycle. Usual Eulerian routing provides deterministic transport delays but has a very low network utilization. With this new strategy the average transport time is much smaller and we can prove a deterministic upper bound of the transport delay, unlike deflection routing which suffers from livelocks. We study the performance guarantees provided by this new algorithm using graph arguments for the ending property and simulations to give some insights for the performance of the algorithm.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133978561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：